ADULT HEAD KIDNEY OF BDELLOSTOMA STOUTI 859 
The arter^^ supplying the glomerulus, in most cases at least, 
is not given off directly from the aorta, but is a branch of an 
intersegmental artery. (This is true also for the arteries supplying 
the glomeruli of the functional kidney.) The artery may divide 
into two or three branches before entering the glomerulus. The 
veins carrying blood away usually pass forward along the central 
duct, branching as they go into a few smaller veins, which in 
the material studied soon became lost. There is no accessary 
agreement between the number of arteries and the number of 
veins, nor between either of these and the number of elements 
which have united to form the glomerulus. 
The glomerulus lies in a cavity formed by the fusion of the 
cavities of the Malpighian corpuscles of the embryo. In cases 
where the gbmeruH remain distinct this fusion may be only partial, 
the different cavities, each occupied by a glomerulus, being con- 
nected with one another by small openings. As a rule the cavit}^ 
is in communication with the pericardial cavity through a variable 
number of apertures, though this is not always true. The aper- 
tures may be very small, or they may be so large that in some 
sections the glomerulus presents the appearance of hanging freely 
in the coelom, like the glomus of a pronephros. 
The cavity is connected with the central duct by short tubules. 
These resemble the other tubules of che head kidney so closely 
that in sections where both occur the two can not be certainly 
distinguished except by tracing them to their termination. The}^ 
project into the cavity containing the glomerulus, where they 
open through nephros comes (fig. 3) ; and in some cases they resem- 
ble the other tubules still more closely by branching and opening 
by from two to five nephrostomes. In a single case a tubule was 
seen to divide into two branches, one of which opened iito the 
cavity of the glomerulus and the other into the pericardial 
cavity. The number of tubules is not always the same as the 
original number of glomeruli. In one instance where there were 
three distinct glomeruli there was but a single tubule, and in 
another case there was no tubule at all. It must be left a ques- 
tion whether these tubules are ciliated, since none of them were 
examined in the living state; but in sections some of them had 
